The Audio Video Coding Standard of China (AVS) is a streamlined, highly efficient video coder employing the latest video coding tools for high definition and high quality video services. AVS provides video coding algorithms that are optimized between absolute coding performance and complexity of implementation. Exemplary AVS applications may comprise broadcast TV, HD-DVD, and broadband video networking.
AVS is a set of standards comprising AVS1-P1 (System for broadcast), AVS1-P2 (Video for HD), AVS1-P3 (Audio, 2 and 5.1 channels), AVS1-P6 (DRM), AVS1-P8 (Video for mobile), and AVS1-P9 (File format). The AVS1-P2 Jizhun profile has been standardized, and other portions of AVS are currently being work on. The AVS1-P2 video codec has a very broad application range that covers various forms of digital compressed video from low bit-rate Internet streaming applications to HDTV broadcast and Digital Cinema applications with nearly lossless coding.
The AVS1-P2 specification defines AVS1-P2 bitstream syntax and specifies a complete process required to decode the bitstream. The AVS1-P2 bitstream is defined as a hierarchy of layers with a video sequence layer as the outermost layer, a picture layer, a slice layer, a macroblock layer and a block layer, accordingly. A start code based bitstream structure has been adopted in the AVS1-P2 specification. Each layer of the AVS1-P2 bitstream begins with a unique start code allowing a decoder's parser to find each layer within the AVS1-P2 bitstream.
The AVS working group provides two reference models of the AVS1-P2 bitstream in the AVS specification version AVS1-P2 Rm52j_r1 and the AVS specification version AVS1-P2, respectively. The reference model of the AVS1-P2 bitstream in the AVS specification version AVS1-P2 Rm52j_r1 is not fully compliant with the AVS specification version AVS1-P2. For example, in direct mode of the AVS specification version AVS1-P2 Rm52j_r1, for the derivation process of, for example, DistanceIndexRef, a current picture is field coded while a backward reference picture is frame coded. This is different from the derivation process in the AVS specification version AVS1-P2.
Further limitations and disadvantages of conventional and traditional approaches will become apparent to one of skill in the art, through comparison of such systems with the present invention as set forth in the remainder of the present application with reference to the drawings.